Highthroughput sequencing from museum samples

BMS Summer Seminar 2024

Saad Arif/Rebecca Whitla

Why are museum specimens valuable for DNA?

A global spatio-temporal record of populations and species, spanning 200 years.

  • useful for studying extinct, extirpated or threatened species/populations

  • building baselines or study long term changes

  • metagenomics

  • vertebrate bias in current musueum-based population genomic studies

What are the issues with historical DNA (hDNA)?

These are mostly similar to ancient DNA (aDNA) but perhaps not as exaggerated.

Contamination Quality

  • Contamination - multiple sources
  • DNA fragmentation - typically 50 bp fragments
  • Deamination at 3’ and 5’ ends
  • low yield
  • Choice of tissue

Images: Raxworthy & Smith 2021, TREE

Dealing with hDNA extraction/library prep

  • Choice of tissue - legs kept curator’s happy, also gave use best libraries
  • Contamination - workspace/tools only for museum samples, run & sequence extraction blanks
  • Deamination - can be fixed fully or partially fixed with commercially available repair enzymes. Dealt with at computational stage.
  • low yield - for our tissue (butterfly legs) a qiagen tissuelyzer prep worked best (further details in Rebecca’s thesis).
  • DNA fragmentation - problematic for illumina library preparation - difficult to separate adapter dimers in library prep, use beads for size selection and titration of adapter to reduce adapter contamination

 

The last days of Aporia crataegi in Britain: evaluating genomic erosion in an extirpated butterfly

Whitla et al. 2024, Mol Ecol

A brief introduction to Aporia crataegi (the black-veined white)

Also extirpated in the Netherlands and South Korea

Images: Todisco et al 2020, Sci. Rep.

The “mysterious” disappearance of Aporia crataegi in Britain

  • Correspondence in the Entomological journals suggest that the butterfly was already disappearing from the western part of its range. Last sighting in Kent 1925 (Pratt 1983, Entomologist’s Rec. J. Var)

  • Record of numerous releases within the 19th century - unclear if resident GB population survived into 20th century.

  • Debate on reasons of extinction:

    • Avian predation
    • fungal, viral or other pathogen
    • Series of unfavorable wet Septembers

“much mystery hangs over this species” - C.G. Barrett, 1893

A genomic perspective on the mysterious disappearcne of Aporia crataegi in Britain

Goals:

  • Were UK populations of A. crataegi distinct from European stock? Did they survive into the 20th Century?

  • Was the decline rapid or fast? For the latter we could expect to see no genomic erosion

    • Compare relative genomic erosion in UK A. crataegi to mainland European pop. Perform same analysis for a demographically stable species (Polyommatus icarus) to provide some context.

  • Is there any evidence for a role of pathogen infection in the extirpation of UK butterflies?

Sampling and Methods: Field sampling

  • Sampled specimens only where collection year and locality was available
  • 1-2 legs per butterfly dissected

Sampling and Methods: Sequencing Library Prep

  • DNA extraction:
    • one or two legs macerated using Tissuelyser II bead mill (Qiagen) and finished with QIAamp DNA Micro kit (Qiagen)
  • Library preparation:
    • NEBNext Ultra II DNA Library Prep Kit using a modified CUT&RUN procotol
    • Exclusion of adapter dimers using SPRIselect beads and titrating primer content
    • 10-12 PCR cycles depending on input template amount
    • dual indices
  • Sequencing:
    • High quality genome assembly and annotation via Darwin Tree of Life
    • individual libraries pooled in equimolar ratio
    • extraction blanks pooled in 1:10 ratio for controls
    • PE150 bp sequencing at Novogene

Sampling and Methods: Data Processing

Images: Kutschera et al. 2022, BMC Bioinfo.

Sampling and Methods: Data Analysis

  • Population Structure - plinkPCA and angsdPCA (for low coverage data)

  • Historical demography - PSMC on highest coverage genomes available

  • Genetic diversity - individual heterozygosity estimated with angsd

  • Inbreeding - Runs of homozygosity determined with plinkPCA

  • Genetic load - predicted LOF, missense and synonymous variants with snpEFF

  • any statistical analysis is either a t-test or when comparing samples with variable coverage, a linear model with coverage covariate.

Results: GB A. crataegi are distinct from European

  • Isolation by distance pattern within GB
  • Evidence of European stock releases in late nineteenth/early 20th century
  • Oldest specimen (1924) in our sampling clusters with GB specimens

Results: Evidence of strong bottleneck at time of GB colonisation

  • Bottleneck signal also present in P. icarus but not as strong, potential recovery
  • Difference in P. icarus demography potenially due to rapid colonisation?

Results: A. crataegi and P. icarus show similar levels of reduction in Genetic diversity

17% reduction in GB A. crataegi and 14% P. icarus relative to EU

Results: Higher levels of inbreeding in GB A. crataegi

Levels of inbreeding similar to those observed in extinct Glaucopsyche xerces (de-Dios et al. 2023, elife)

Results: Increase in realised genetic load in GB A. crataegi

Increase in homozygous synonymous and missense mutations

Conclusions

  • Resident GB A. crataegi survived into 20th century, however 2/4 20th century samples are likely of EU ancestry

  • A. crataegi underwent severe bottlneck at time of colonisation and never seems to have recovered. Increases in homozygosity suggest a spasmodic rather than instant decline in the 19th-20th century

  • Increase in homozygous missense mutations may have made population more prone to extinction

Broader signficance:

  • Consistent differences in genomic erosion suggest these metrics could be helpful to monitor or identify threatened insect populations/species with a small number of individuals.

  • Our DNA extraction + Library prep works well:

    • use of legs keeps curators mostly happy

    • ≥50% coverage at depth of ≥5X for ~175 yo specimens, ≥90% coverage at depth of ≥5X for ~120 yo specimens

Further work:

  • What was the ultimate cause of the regional extinction?

    • at least two other Butterfly species disappeared from the UK around the same time

Further work: Metagenomics

  • Aspergillus sp. are known to be entomopathogenic but they are quite common and potentially might be perimortem or postmortem contamination

  • more work needed

The future: Functional Genomics in museum specimens

  • Genome-wide methylation is possible from veretbrate museum specimens

Acknowledgements

Rebecca Whitla (PhD student)

Collaborators:

Prof Tim Shreeve

Dr. Casper Breuker

Dr. Korneel Hens

Geoff Martin (NHM, London)

Dr. James Hogan (OUMNH, Oxford)

Other:

Kornad Lohse and Darwin Tree of Life


Funding:

Nigel Groome Studentship to RW by Dept. of BMS, Oxford Brookes University


More Info:

Lab: saadarif.github.io

the ‘scrolltelling’ experience: saadarif.github.io/BVW-LastDays

email sarif@brookes.ac.uk

What is Genomic Erosion and why it matters?

  • The loss of genetic fitness in small populations due to a shrinking ‘gene pool’, but measured by genetic variation

  • The promise of conservation genomics lies in the link between genomic variation and genetic fitness, as the latter is potentially measurable

  • lack of knowledge is more acute for invertebrates! Evaluating genomic erosion in inverterbates with differing conservation status could improve our understanding and application of conservation genomics

Images: Bosse & Loon 2022, Fron. Gen.

Results: overall increase in shared derived synonymous and missense mutations n GB A. crataegi

Still cannot detect difference in LOF mutations